The Amateur Seflight Association (ASA) of Houston, Texas has a mission: Promote space-related activities in such a manner that safely challenges the technical limits of amateur space flight. Breaking the world record for amateur rocket altitude would help accomplish this goal, but the ASA needs more than a bottle of liquid oxygen, a book of matches and a really long ruler. Given a payload of opponity, Nicolaus Radford, Chief Avionics Engineer for ASA went ballistic with the RCM2300 Microprocessor Core Module.
"I got into Rabbit about 3 years ago, basically just looking for an embedded processor that would fit the form factor and capability I was after. I came across the Rabbit Semiconductor web site, bought one, and have not looked back since." From there, Radford described ASA most recent rocket endeavor with Rabbit power. "The rocket that we launched was about 18.5 ft. in length and 9 inches in diameter. It pulled about 8.5 G on takeoff, broke the sound barrier, and flew up to about 12000 ft. The rocket had video, telemetry, GPS, full dynamical analysis of the rocket, all real-time," stated Radford. "What we were having to do on the rocket was measure all these vehicle parameters including acceleration, velocity, pressure, and position. We built a custom data acquisition board that had all the components on it. We were using the Rabbit and an I2C bus to read and store data in the Rabbit Flash. At the same time, the Rabbit was also serially interfacing to a terminal node controller and a GPS unit on 2 serial ports. The Rabbit would run out on the I2C bus and read all these different sensors around the rocket and then it would format the data and send it to the terminal node controller, which would then format the data for packet radio and send it down on a HAM radio frequency. Intermittent to this were GPS readings being sent via our telemetry band to the ground as well."
Concerning the ease of designing, programming, and communicating with the RCM2300: "The communications bus was an I2C protocol. We absolutely loved it! It was very simple to string up multiple I2C controllers and talk to them to get all the data back. It was great because on a long rocket you have sensors in the nose cone and the body, and it was nice to interface with these sensors with 2 wires instead of a string of wires. We used 2 of the 4 serial ports on the Rabbit to interface to a terminal node controller, which interfaced to a HAM radio. We also interfaced to a GPS receiver. When you look at how small the RCM2300 is compared to what it does, I do not think there is anything on the market that can beat it. You have got 40 I/O pins, 4 serial ports, Flash memory to store data-you do not need any external interface to it to run our system except the core module. The size space savings is just tremendous for the capability. And it has got relatively low power consumption-110 mA at 5 V is pretty small!"
In regards to the usage of software in the project, critical components of Dynamic C helped Radford get to the launch pad. "We probably could not have done the project in the timeframe and ease that we did if we did not have the costate construct. It was just phenomenally easy to use. It basically simulates a multi-threaded environment with hardly any overhead what so ever. The way Dynamic C time-slices and the way the event control is setup is just amazing. If you were to try to write out and code out to simulate a multi-thread yourself, you would be bogged down in months of coding. I do not think as a programmer I could make a better utilization of processing time than what the Rabbit did using the costatement. And I have yet to see this anywhere else."
The bottom line for development time, ease, and risk reduction is what Rabbit customers have voiced as the most important factors for design wins. Radford elaborated on his own experience: "Our development time to completely get the system the way we wanted it was eight months. Had we not had the development kit and protoboard to easily interface the programming pins of the Rabbit, had we used some other processor where we would continuously have to fabricate our own ways to interface to the processor, it would not have been nearly as fast. It was really nice that the pins were broken out to individual headers on the protoboard-that made development tremendously easy. The ready-to-go protoboard surface mount pads helped significantly as well. My best guess is that our development time would have been twice as long if we had not gotten these benefits in the development kit."
During flight, the rocket performed perfectly. It was not until the way down that ASA had a scare in the form of a failed main parachute. "Upon landing, our data told us that we took 61 G on impact. Even though the backup parachute successfully protected the entire rocket, it was pretty impressive that the Rabbit withstood the shock." Radford laughed.
So what is in store for the ASA in the future? Radford explains, "[This launch] is all in preparation for a space shot we are going to be attempting early next year of a 40 ft. liquid motor vehicle. It will enter space, reach 400,000 ft. and return. We will have many Rabbits on that flight." Radford continued, "Our next design where we actually have throttle control ability, the Rabbit will be used as a motor controller. We will have multiple Rabbits; one of which will be used for liquid motor throttling and data acquisition of the liquid motor. We will also use the Rabbit to analyze power usage and to turn things on and off as they are needed to conserve power."
"We have also already started a product spin-off from our main vehicle system. It is a data acquisition system aimed toward the high-powered rocketry community. The rocketry community is all about size: how small can you get it? The first thing that struck us about the RCM2300 was how small it was for what it did. That was extremely important to us because we did not have a lot of space to work with. The benefit over say, a PIC processor, was that the Rabbit has all the extra overhead, the Flash memory interfaces that you need to add onto a PIC and 3 ti the processing power. We believe that the Rabbit will give us a product we can take to market that will be superior to anything currently available for a data acquisition system in high-powered rocketry applications."